Application of aggregate analysis for product design quality using QFD model and TOPSIS/OFD modelio ir topsis metodo kompleksines analizes taikymas gaminiu projektavimo kokybei gerinti.
Pang, Jihong ; Zhang, Genbao ; Chen, Guohua 等
1. Introduction
Design quality has a great influence on the success of a product.
And the product design quality can be evaluated by various mature in
product development phases [1]. On the other hand, the design quality
should be consistently on product design target. More importantly, for
the purpose of improving the level of product design quality, the
suitable concept design is crucial [2]. In order to facilitate the
process of transforming product design quality objectives into effective
actions, product quality would be considered by many engineering design
requirements (EDRs) [3]. In today's rapidly evolving world of
manufacturing industry, design quality of mechanical and electrical
products is more and more important [4]. Consequently, the aggregate
analysis of product design optimization is a critical part for product
quality during the design scheme decision processes [5].
An optimized design is the first step in product development, and
many EDRs need to be considered. There are different product design
quality models available in the present literature. Among lots of
analysis techniques, Quality function deployment (QFD) model is
particularly famous for its successful applications of transforming
customer satisfaction into design stage. Technique for order preference
by similarity to ideal solution (TOPSIS) method has been used to assist
the decision making in many fields. In addition, the integration of QFD
and TOPSIS can achieve good performance on many tasks. Moreover, product
design quality researchers have become increasingly aware of the
decision problems. The aim of this paper is to clarify the aggregate
analysis by using QFD and TOPSIS and solve the problem of quantitative
decision for product design quality.
The remainder of this paper is organized as follows. In Section 2,
the proposed method for product design quality project selection problem
is presented. Section 3 introduces the novel decision model based on QFD
and TOPSIS. An application of aggregate analysis for design quality of a
gas turbine is discussed in Section 4. The last section summarizes the
findings of this research and closes with directions for further
research.
2. The aggregate analysis method
In order to improve the product design quality, the customer needs
should be accurately transformed into engineering technical requirements
for the product design. The aggregate analysis model of this paper is
established on the basis of QFD model and TOPSIS method. The proposed
aggregate analysis model is illustrated in Fig. 1. The initial phase of
the proposed model defines customer requirements (CRs) for the product
design quality. Then, the house of quality (HoQ) has been integrated
with QFD methodology in order to transformed CRs into EDRs. And the
importance of EDRs can be accurately determined by using HoQ. Based on
the data of EDRs provided by product design quality, the optimization
design is easily identified based on TOPSIS method. So, the best concept
design can be acquired with the max design quality.
[FIGURE 1 OMITTED]
3. Synthesis analysis model
Traditional design statistics data show some specifications and
characteristics of a product structure, and only the product design
quality that could be measured accurately is utilized. The systematic
decision support approach integrates with QFD and TOPSIS for product
design quality are described as follows.
3.1. QFD analysis model
The concept of QFD was first initiated by Akao in 1966, which is a
customer-driven design method [6]. The QFD is a useful tool for product
design, development and management. QFD meet the needs of customers and
carries out a competitive analysis for design indicators.
In order to translate the voice of the customers through the
various stages of product design, each translation matrix called house
of quality (HoQ) is applied, as is shown in Fig. 2. The requirements
relationships between customers needs and technical requirements are
adjusted by HoQ. After obtaining what the customers want and need, the
HoQ was applied to translate the customer needs into the measurable
engineering characteristics. Recently, the initial structure and
principles of QFD have been successfully used to manage design
information and assist decision-making in product development process.
HoQ is an effective method to ensure attention for comfort in the
product design process. The standard HoQ includes customer requirements
(WHATs), engineering design requirements (HOWs) and relationship matrix
of WHATs. Firstly, the customer requirements (CRs) are collected through
market research by a design team. Next, the correlation between EDRs and
CRs is calculated by expert practice advice. Finally, the importance of
EDRs which affects CRs is indicated in the engineer language.
In practice, QFD is widely used as the most important technical
tool to translate CRs into product technical requirements of new product
development. QFD was applied to deal with product development for
achieving the max product quality by meeting customer request [7]. With
the purpose of gaining the best customer satisfaction, QFD was used to
solve the correlation triangle problem of converting CRs into
engineering characteristics (ECs) [8]. QFD integrated with robust design
made a design solution for multiple optimization problems of product
design [9].
3.2. TOPSIS decision method
The TOPSIS method is firstly proposed by Hwang and Lin in 1987
[10]. In general, TOPSIS has two major functions: the one is to
calculate the longest distance from negative ideal solution; the other
is to choose the optimization alternative which has the shortest
distance from ideal solution. In case of analysis for decision problem,
TOPSIS is an effective and practicable method used for rank ordering
schemes by preference.
TOPSIS method has been successfully applied to solve multicriteria
decision making problem in various industrial field. The multiattribute
decision making model based on TOPSIS was arranged for the disposal of
decision problem of logistic information technology [11]. TOPSIS was
used to manage competitive benchmarking in product design process [12].
TOPSIS integrated with other methods were developed to deal with
multipurpose reactive power compensation problem [13].
In this paper, the TOPSIS method was used to analysis decision
process for product design quality because the concept is available and
reasonable. The steps in the general TOPSIS process can be described as
follows.
Step 1: Construct the normalized decision matrix. The vector
combines the concepts of decision matrix in the following expression
[MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII.] (1)
where [g.sub.mk] represents the mth alternative for the nth
attribute. And the normalized decision matrix can be calculated by
[MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII.] (2)
Step 2: Find the weighted normalized matrix. First, the set of
importance weights of [w.sub.j] are developed by QFD model. Then, the
weighted normalized matrix can be constructed as follows
[??] = [[[V.sub.ij].sub.mxn] = (3)
where [MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII.].
Step 3: Identify ideal and antiideal solution. The ideal solution
([V.sup.*]) is shown in the following
V = max{[v.sup.*.sub.1], [v.sup.*.sub.2], ..., [v.sup.*.sub.j]}, j
= 1,2, ..., n. (4)
Similarly, antiideal solution [V.sup.-] is determined as
[V.sup.-] = min{[v.sup.*.sub.1], [v.sup.*.sub.2], ...,
[v.sup.*.sub.j]}, j = 1,2, ..., n. (5)
Step 4: Develop the distances between each alternative. The
distances of each alternative from ideal solution can be calculated by
the equation given below
[MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII.] (6)
The distances for antiideal solution are calculated as
[MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII.] (7)
Step 5: Calculate the closeness coefficient. The ranking order of
all alternatives by the sum of the distance to the ideal solution is
[C.sup.*.sub.i] = [d.sup.-.sub.i]/([d.sup.*] + [d.sup.-.sub.i]) (8)
Step 6: Rank the alternatives. The preference order can be decided
by Eq. (8), which is close to the ideal solution and far from the
antiideal solution.
Step 7: Recommend the best alternative. The preferred alternative
is the one with the maximum ratio of [C.sup.*.sub.i].
4. An empirical application
The described methodology is applied to analysis product design
quality by using QFD model and TOPSIS. The experiment was basically
setup upon the decision of the design quality for a gas turbine. The
number of CRs is heavily dependent on the consumer satisfaction. Here 5
CRs were selected: quality assurance ([CRs.sub.1]), reliability
([CRs.sub.2]), maintainability ([CRs.sub.3]), indemnificatory
([CRs.sub.4]), availability ([CRs.sub.5]). In order to response to
consumer requirements, EDRs are applied based on QFD model. In addition,
the EDRs were consulted to construct relationship matrix based on Power
([EDRs.sub.1]), Engine Speed ([EDRs.sub.2]), Max Service Life
([EDRs.sub.3]), Compression Ratio ([EDRs.sub.4]), Noise ([EDRs.sub.5]),
Exhaust Pollution (EDRs6, (NOx, COx)) and Combustion Efficiency (EDRs7).
The information and data for HoQ are depicted in Table 1.
The next step is to determine the EDRs of design quality for a gas
turbine according to the CRs. For example, the importance value of
[EDRs.sub.1] is calculated by using the correlation of HoQ as following
[EDRs.sub.1] = 0.35x3 + 0.15x9 +0.10x1+0.25x3+0.15x9 = 4.60.
Similarly, the other EDRs are determined as: [EDRs.sub.2] = 6.40,
[EDRs.sub.3] = 2.90, [EDRs.sub.4] = 3.10, [EDRs.sub.5] = 4.40,
[EDRs.sub.6] = 3.60, [EDRs.sub.7] = 4.90. Then, the importance can be
obtained as: W = (0.1538, 0.2140, 0.0970, 0.1037, 0.1472, 0.1204,
0.1639). After translating CRs into EDRs, the data comparison of four
solutions is completed in Table 2.
Then, by applying Eq. (2), the normalized decision matrix is
calculated for each alternative. Next, we calculated weighted normalized
decision matrix, and Eq. (3) is applied to calculate the total matrix.
After that, the ideal solution is calculated by using the data of EDRs
via Eq. (4). [V.sup.*] = (0.0837, 0.1204, 0.0544, 0.0448, 0.0663,
0.0428, 0.0736). Similarly, the anti-ideal solution is computed as
follows: V= (0.0698, 0.0928, 0.0415, 0.0565, 0.0858, 0.0770, 0.0935).
Next, using the data in Table 2 and Eq. (6), we can calculate the
distances from the ideal solution as
[d.sup.*.sub.1] = 0.0336, [d.sup.*.sub.2] = 0.0289, [d.sup.*.sub.3]
= 0.0329, [d.sup.*.sub.4] = 0.0390
The distances for antiideal solution can be obtained in the same
way using Eq. (7)
[d.sup.-.sub.1] = 0.0374, [d.sup.-.sub.2] = 0.0374, [d.sup.-.sub.3]
= 0.0447, [d.sup.-.sub.4] = 0.0276
In the next step, the closeness coefficient to the ideal solution
is given. Finally, the aggregate analysis results for design quality are
shown in Table 3. According to the analysis, solution No.3 is the best
performer among four schemes. The final ranking obtained by the proposed
method was totally in accordance with the intuitive preference of design
quality for the gas turbine.
5. Conclusions
The product design quality selection problem formulated as
multiobjective optimization problem with competing amount of quality
indicators. This paper has proposed a new integrated QFD model and
TOPSIS method for product design quality in the manufacturing
performance selection. In order to translate CRs into product technical
requirements, the appropriate criteria weights of EDRs are obtained
using QFD model. Then, we developed the HoQ model for dealing with
various types of uncertain information of CRs.
In addition, the TOPSIS approach was fairly used to denote the
level of design solution responding the performance difference. After
the weights are obtained by QFD, the aggregate performance of each
alternative is easier to achieve. The methodology solves the ambiguity
of the comparison process by using the relative position of ideal and
antiideal solution. For the comparison of all solution, we have selected
the best alternative according to the aggregate analysis results.
The proposed aggregate analysis model has practical application as
the empirical test showed in the case of design quality selection
problem of a gas turbine. Furthermore, the proposed method is also used
to solve other op timization problems in various industries. As a future
step to this paper could be the comparison of the proposed approach to
other multiple criteria group decision-making methods.
Acknowledgment
This work was supported by National High-Tech. R&D Program,
China (No. 2009AA04Z119), the National Natural Science Foundation, China
(No. 50835008), the National Major Scientific & Technological
Special Project "High-grade CNC and Basic Manufacturing
Equipment", China (No. 2009ZX04014-016; No. 2009ZX04001-013; No.
2009ZX04001-023; No. 2010ZX04014-015).
Received March 15, 2011
Accepted December 05, 2011
References
[1.] Cikotiene, D.; Bargelis, A. 2009. Research of quality impact
to the product design properties and characteristics, Mechanika 5(79):
63-67.
[2.] Povilionis, A.; Bargelis, A. 2010. Structural optimization in
product design process, Mechanika 1(81): 66-70.
[3.] Govindaluri, S.M.; Cho, B.R. 2007. Robust design modeling with
correlated quality characteristics using a multicriteria decision
framework, International Journal of Advanced Manufacturing Technology
32(5-6): 423433.
[4.] Freiesleben, J. 2010. Proposing a new approach to discussing
economic effects of design quality, International Journal of Production
Economics 124(2): 348-359.
[5.] Allahbakhsh, H.R.; Saemi, J.; Hourali, M. 2011. Design
optimization of square aluminium damage columns with crashworthiness
criteria, Mechanika 17(2): 187-192.
[6.] Mrad, F. 1997. An industrial workstation characterization and
selection using quality function deployment, Quality and Reliability
Engineering International 13(5): 261-268.
[7.] Reich, Y.; Levy, E. 2004. Managing product design quality
under resource constraints, International Journal of Production Research
42(13): 2555-2572.
[8.] Sharma, J.R.; Rawani, A.M. 2007. Ranking customer requirements
in QFD by factoring in their interrelationship values, Quality
Management Journal 14(4): 53-60.
[9.] Kovach, J.; Cho, B.R. 2008. Solving multiresponse optimization
problems using quality function-based robust design, Quality Engineering
20(3): 346-360.
[10.] Abo-Sinna, M.A.; Amer, A.H. 2005. Extensions of TOPSIS for
multi-objective large-scale nonlinear programming problems, Applied
Mathematics and Computation 162(1): 243-256.
[11.] Kahraman, C.; Ates, N.Y.; Cevik, S.; Gulbay, M.; Erdogan,
S.A. 2007. Hierarchical fuzzy TOPSIS model for selection among logistics
information technologies, Journal of Enterprise Information Management
20(2): 143-168.
[12.] Lin, M.C.; Wang, C.C.; Chen, M.S.; Chang, C.A. 2008. Using
AHP and TOPSIS approaches in customer-driven product design process,
Computers in Industry 59(1): 17-31.
[13.] Azzam, M.; Mousa, A.A. 2010. Using genetic algorithm and
TOPSIS technique for multiobjective reactive power compensation,
Electric Power Systems Research 80(6): 675-681.
Jihong Pang *, Genbao Zhang **, Guohua Chen **
* College of Mechanical Engineering, Chongqing University,
Chongqing 400044, China; School of Mechanical Engineering, Wenzhou
University, Wenzhou 325035, China, E-mail: pangjihong@163.com
** College of Mechanical Engineering, Chongqing University,
Chongqing 400044, China, E-mail: genbaozhang@163.com
Table 1
Original HoQ for design quality of gas turbine
Weight Power Engine Max Compression
speed service ratio
life
Quality assurance 0.35 3 9 1 3
Reliability 0.15 9 3 9 1
Maintainability 0.10 1 1 3 3
Indemnificatory 0.25 3 9 3 1
Availability 0.15 9 3 1 9
Noise Exhaust Combustion
pollution efficiency
Quality assurance 9 3 9
Reliability 3 1 3
Maintainability 1 3 9
Indemnificatory 1 3 1
Availability 3 9 1
Table 2
Data for four solutions with seven EDRs
Engineering design requirements (EDRs)
No. [EDRs.sub.1] [EDRs.sub.2] [EDRs.sub.3] [EDRs.sub.4]
kW rpm hour scale
1 75 85000 55000 5.5
2 80 75000 45000 4.6
3 90 65500 42000 5.8
4 85 75500 53000 5.3
No. [EDRs.sub.5] [EDRs.sub.6] [EDRs.sub.7]
dB ppm %
1 60 15 23.5
2 75 12 18.5
3 58 10 19.5
4 63 18 20.5
Table 3
Aggregate analysis results for design quality
Solution No. d * d C * Sort
1 0.0336 0.0374 0.5263 3
2 0.0289 0.0374 0.5643 2
3 0.0329 0.0447 0.5761 1
4 0.0390 0.0276 0.4143 4
Fig. 2 Standard house of quality
Correlation between EDRs
[EDRs.sub.1] [EDRs.sub.2] ...
CRs Weight
[CRs.sub.1] [W.sub.1] [R.sub.11] [R.sub.12] ...
[CRs.sub.2] [W.sub.2] [R.sub.21] [R.sub.22] ...
... ... ... ... ...
[CRs.sub.m] [w.sub.m] [R.sub.ml] [R.sub.m2] ...
The importance of EDRs
[EDRs.sub.n]
CRs
[CRs.sub.1] [R.sub.1n]
[CRs.sub.2] [R.sub.2n]
... ...
[CRs.sub.m] [R.sub.mn]
